9:30 - 10:30 Keynote talk: Symmetry/Structure Detection and Preservation in 3D Geometry,
Niloy Mitra (KAUST)
Self-similarity or symmetry is commonly observed in many natural and
man-made objects. As we enter the age of easily accessible large
collections of 3D data, such global understanding of the underlying
structure of data becomes more relevant and useful. In this talk I
will summarize our recent contributions towards detecting symmetries
and structures in 3D geometry. I will briefly describe an algorithm
for detecting partial and approximate symmetries in 3D shapes, and
then propose an extension which discovers underlying structure among
repeated elements in the scenes.
State-of-the-art deformation tools fall short at preserving
characteristic features and global structure in man-made objects. We
introduce iWires, a novel analyze-and-edit approach based on the
argument that man-made models can be distilled using a few special 1D
wires and their mutual relations. In our approach, prior to editing,
we perform a light-weight analysis of the input shape to extract a
descriptive set of wires. Editing the object by modifying the
intelligent wires leads to a powerful editing framework that retains
the original design intent and object characteristics.
Shape similarity and retrieval
- 10:30 - 10:55 Content-aware image resizing by quadratic programming
- 10:55 - 11:20 Efficient retrieval of deformable shape classes using local self-similarities
Inverse problems
- 11:40 - 12:05 On reconstruction of non-rigid shapes with intrinsic regularization
- 12:05 - 12:30 Detailed body shapes from flash photographs
14:00 - 15:00 Keynote talk: Modeling Deformable Surfaces from Single Videos,
Pascal Fua (EPFL)
Without a strong model, 3--D shape recovery of non-rigid surfaces from
monocular video sequences is a severely under-constrained problem.
Prior models are required to resolve the inherent ambiguities. In our
work, we have investigated several approaches to incorporating such
priors without making unwarranted assumptions about the physical
properties of the surfaces we are dealing with.
In this talk, I will present these approaches and discuss their relative
strengths and weaknesses. I will also demonstrate that they can be
incorporated into effective algorithms that can capture very complex
deformations.
Deformable shapes and image alignment
- 15:20 - 15:45 Probabilistic constrained adaptive local displacement experts
- 15:45 - 16:10
Markov chain Monte Carlo shape sampling using level sets (Best paper according to audience vote)
- 16:10 - 16:35
Fast nonrigid mesh registration with a data-driven deformation prior
- 16:35 - 17:00
Non-rigid registration between color channels based on joint-histogram entropy in subspace
9:30 - 18:00 Poster session
-
Shape Google: a computer vision approach to invariant shape retrieval
-
Learning shape metrics based on deformations and transport
-
Unsupervised learning of human body parts from video footage
-
Learning varying dimension radial basis functions for deformable image alignment
-
Uncalibrated non-rigid factorisation with automatic shape basis selection
-
Integrating contour and skeleton for shape classification
-
Online active feature model for lip tracking
-
Effective and efficient interpolation for mutual information based multimodality elastic image registration
-
Bending invariant meshes and application to groupwise correspondences
-
Joint estimation of deformable motion and photometric parameters in single view video
-
A phase field higher-order active contour model of directed networks
|